Rules for Sorting Ideographic Characters
There are several traditional methods for sorting ideographic characters:
By radical, and then by the number of strokes required to draw the radical, in increasing order
By the number of strokes required to draw the character, and then by the order of the radicals
By pronunciation (note that many ideographs have more than one pronunciation)
By code-point value
Far East editions of Windows support the sort methods listed in the following table. The sort method used depends on the locale specified in the sort call. By default, each Far East edition supports only the Unicode sort order and the sort order or orders for the default system locale. The user can, however, install sort tables for additional languages via Control Panel.
| Locale | Sort Order |
| Any locale that is not Chinese, Japanese, or Korean | By Unicode code-point value. Ideographic characters in Unicode are grouped according to the number of strokes required to draw the radical and then by the number of strokesrequired to draw the rest of the character |
| Japanese | In Shift-JIS order. Roman and kana characters in Shift-JIS are sorted phonetically, and kanji characters are sorted by radical. |
| Traditional Chinese | By the number of strokes required to draw the character. |
| Simplified Chinese | In phonetic order. This is the default for the Simplified Chinese locale. |
| or | |
| By the number of strokes required to draw the character. This is provided as an alternative sort method. | |
| Korean | In phonetic order. Hanja and hangul characters are intermixed. |